Wilcoxon Signed RANK TEST
Uploaded by
AnkushWilcoxon Signed RANK TEST
Uploaded by
AnkushWilcoxon Signed-Rank Test (10-mark
answer)
Definition
The Wilcoxon Signed-Rank Test is a non-parametric statistical test used to compare two
related (paired) samples.
It is the non-parametric alternative to the paired t-test.
It evaluates whether the median difference between paired observations is zero.
When is it used?
Use the Wilcoxon Signed-Rank Test when:
1. You have paired or matched observations
o e.g., pre-test vs post-test, left vs right eye, same patient on two drugs
2. Outcome variable is continuous or ordinal
3. Data is not normally distributed
4. You want to test median difference rather than mean difference
Clinical Scenario
A clinician wants to test whether a new antihypertensive drug significantly reduces systolic
blood pressure (SBP) after 4 weeks.
He measures SBP in 8 patients before and after treatment.
Data (paired):
Patient SBP Before SBP After
1 150 140
2 160 150
3 145 140
4 170 160
5 155 150
6 148 147
7 162 150
8 158 155
Data is not normally distributed, so Wilcoxon Signed-Rank Test is chosen.
Steps in Wilcoxon Signed-Rank Test
1. State hypotheses
Null hypothesis (H₀):
The median difference between paired values is zero (drug has no effect).
Alternative hypothesis (H₁):
The median difference is not zero (drug changes BP).
2. Calculate differences
Difference=(After−Before)
Patient Diff Absolute Diff
1 -10 10
2 -10 10
3 -5 5
4 -10 10
5 -5 5
6 -1 1
7 -12 12
8 -3 3
(All differences are negative, indicating BP dropped.)
3. Rank the absolute differences
Abs Diff Rank
1 1
3 2
5 3.5
5 3.5
10 6
10 6
10 6
12 8
(Tied values get average ranks.)
4. Assign signs back to the ranks
Since all differences were negative, all ranks are given negative signs.
Patient Diff Abs Diff Rank Signed Rank
1 -10 10 6 -6
2 -10 10 6 -6
3 -5 5 3.5 -3.5
4 -10 10 6 -6
5 -5 5 3.5 -3.5
6 -1 1 1 -1
7 -12 12 8 -8
8 -3 3 2 -2
5. Calculate sums of positive and negative ranks
Sum of positive ranks (T⁺): 0 (no positive ranks)
Sum of negative ranks (T⁻):
−6−6−3.5−6−3.5−1−8−2=−36
6. Choose the test statistic
Test statistic = smaller of |T⁺| and |T⁻|
W =0
7. Compare with critical value / p-value
For n = 8, two-tailed critical value at α = 0.05 ≈ 3.
Since
W =0<3
→ Result is statistically significant.
Interpretation
The Wilcoxon Signed-Rank Test shows a significant reduction in systolic blood pressure
after treatment with the new drug.
Therefore, the new drug effectively lowers SBP.
Advantages
Does not require normal distribution
Suitable for small sample sizes
Uses ranked differences, reducing influence of outliers
Limitations
Less powerful than paired t-test if data is normal
Cannot quantify size of effect as precisely as parametric tests
Conclusion (2-line exam summary)
The Wilcoxon Signed-Rank Test is a non-parametric test for paired samples used to
compare median differences. In the clinical scenario, it showed a significant reduction in
blood pressure after using the new antihypertensive drug.
